A. C. Watts, A. Morris, C. M. Robinson From New Royal Infirmary, Edinburgh, Scotland A. C. Watts, BSc, MRCS(Ed), Orthopaedic Specialist Registrar A. Morris, MBChB, Foundation Year Trainee
C. M. Robinson, FRCS(Tr & Orth), Consultant Orthopaedic Surgeon
Department of Orthopaedics New Royal Infirmary, Little France, Edinburgh EH16 4SU, UK.
Correspondence should be sent to Mr A. C. Watts; e-mail: email@example.com
©2007 British Editorial Society of Bone and Joint Surgery doi:10.1302/0301-620X.89B4. 18284 $2.00
J Bone Joint Surg [Br] 2007;89-B:510-15. Received 16 June 2006; Accepted after revision 11 December 2006
Fractures of the distal humeral articular surface which do not involve the medial and lateral columns are often more extensive than is apparent from plain radiographs. This
retrospective study describes the epidemiology of this injury using modern classification systems and compares pre-operative radiography with operative findings. The study group included 79 patients with a mean age of 47 years (13 to 91). The annual incidence was 1.5 per 100 000 population, and was highest in women over the age of 60. The majority of the fractures (59; 75%) were sustained in falls from standing height. Young males tended to sustain more high-energy injuries with more complex fracture patterns. In 24% of cases (19) there was a concomitant radial head fracture. Classification from plain radiographs often underestimates the true extent of the injury and computed tomography may be of benefit in pre-operative planning, especially in those over 60 years of age.
Fractures of the articular surface of the capi-tellum or trochlea are a distinct, complex subgroup, distinguished from single and bi-column fractures by the lack of significant metaphyseal extension of fracture lines above the olecranon fossa. These injuries are uncommon, but with an ageing population have assumed an increased importance. Since first described in 1841 by Cooper,1 they have
traditionally been viewed as isolated frac-tures of the capitellum, but are frequently more complex, involving a greater part of the distal humeral articular surface. There are numerous series discussing their manage-ment,2-7 but to our knowledge no description of their epidemiology.
The most widely used classification8 is
descriptive and derived from plain radiographs (Fig. 1). More recently, Ring, Jupiter and Gulotta9 described a classification based on operative findings, arguing that the severity of most fractures is underestimated (Fig. 2). However, this system specifically excludes isol-ated capitellar fractures and those involving both metaphyseal columns.
The aim of our study was to present the epidemiology of distal humeral articular frac-tures and to compare the radiological extent of the injury with operative findings, using a modern classification in a well-defined popu-lation.
Patients and Methods
We reviewed the medical records and radio-graphs of all patients attending our unit between 1995 and 2004 with a diagnosis of fracture of the distal humeral articular surface. The hospital is the sole provider of trauma care to a well-defined population. Patients under 13 years of age and those with distal humeral frac-tures involving one or both metaphyseal col-umns were excluded from the study, resulting in a study group of 79 patients (24 male, 55 female) with a mean age of 47 years (13 to 91). Demographic data were collected prospec-tively in a trauma database which maintained details of age, gender, occupation at the time of injury, mechanism, side injured, state of the soft tissues and associated bone or joint injury. The mechanism of injury was grouped as a fall from standing, fall from a greater height, road traffic accident, sports related accident or assault/direct blow. The occupation of the patients was recorded as professional, clerical, manual, domestic, student, retired or unem-ployed.
Pre-operative radiological review. From antero-posterior and lateral radiographs, fractures were grouped according to the McKee modific-ation of the Bryan and Morrey classifica-tion.8,10 Type I (Hahn-Steinthal) is a coronal
shear fracture with one large capitellar frag-ment (Figs 1a and 1b);11,12 type II osteo-chondral lesions of the capitellum are known
by the eponym Kocher-Lorenz (Figs 1c and 1d);13,14 type III
fractures include all comminuted fractures of the capitel-lum.8 McKee et al10 added a fourth pattern, noting that in some cases the Hahn-Steinthal11,12 fracture extends medi-ally in the coronal plane to include the lateral half of the trochlea. This can be recognised by the double arc sign10 on a plain lateral radiograph (Fig. 3). The radiographs were classified independently by two orthopaedic surgeons (ACW, CMR) who were blinded. The level of agreement was assessed by the kappa value. Where there was a differ-ence of opinion, agreement was by discussion and consen-sus. Associated radial head fractures were classified according to Mason15 as type I (undisplaced), type II (dis-placed simple) and type III (comminuted).
Operation note review. Two observers (AM, ACW) who were blinded to the outcome of the radiological
classifica-tion, independently classified the fractures as described in the operation notes according to Ring et al9 (Fig. 2). Again
the level of agreement was assessed by the kappa value. Statistical analysis. The demographic data regarding age, gender and injury patterns were analysed using SPSS for Windows version 13.0 (SPSS Inc., Chicago, Illi-nois). The chi-squared test compared ordinal categorical data with a significance level set at 5%. Assuming the operative findings to be an accurate estimate of the true injury pattern, we assessed the sensitivity, specificity, and positive and negative predictive values of pre-oper-ative radiographs to foretell whether fractures extended beyond the capitellum.
Descriptive radiological classification of a) and b) type 1 Hahn-Steinthal11,12 fracture and c) and d) type II Kocher-Lorenz13,14 fracture.
Ring classification of distal humeral articular fractures showing a) type I, a single articular fragment consisting of the capitellum and the lateral portion of the trochlea, b) type II, associated fracture of the lateral epi-condyle, c) type III, impaction of the metaphyseal bone behind the capi-tellum, d) type IV, fracture of the posterior aspect of the trochlea, e) type V, fracture of the medial epicondyle9 (C, capitellum; T, trochlea; ME,
medial epicondyle; LE, lateral epicondyle).
Fig. 2e Fig. 2d Fig. 2c Fig. 2b Fig. 2a
Epidemiology. The annual incidence of distal humeral artic-ular fracture in this largely urban population was 1.5 per 100 000, with a marked predominance in females and a bimodal distribution with respect to age (over 80 and under
20 years; Fig. 4). In females, the incidence over the age of 60 was twice that between 20 and 59. The incidence in males was much lower, with a unimodal age pattern and a peak incidence under 19 years decreasing with age and becoming rare over the age of 35. The mean age of fracture for males and females was 27.5 years (14 to 62) and 55.1 years (13 to 91), respectively. Their occupations are shown in Table I and the mechanism of injury in Table II.
The right elbow was injured in 29 patients (37%), the left in 50 (63%). There were significant differences in the mechanism of injury between the genders (chi-squared test, p = 0.002). In males this was typically more high-energy, with a quarter being sustained in motor vehicle accidents; only 11 of 22 men (50%) sustained their injury as a result of a fall from standing height, compared with 91% (48 of 53) in females. In four patients the mechanism of injury had not been documented.
Patterns of injury (Tables III to VII). With regard to pre-operative radiographs, the kappa value was 0.74, and for the operation note review it was 0.87. These results indi-cate a very good level of agreement between the two observers.
Any differences in the radiological classification of capitellar fractures between males and females (Table III) did not reach statistical significance (chi-squared t-test, p = 0.2).
Type I (Hahn-Steinthal). This was the most common frac-ture on pre-operative radiographs, accounting for nearly half of the injuries (47%; 37 of 79). The majority were sustained by a fall from standing height. At operation, 28 of the 37 involved the capitellum only and could not be classified using the Ring system (Table III). Nine patients (Table V) had fractures more extensive than suggested by the pre-operative radiographs and extended onto the tro-chlear ridge. Of these, three involved only the capitellum and trochlear ridge, five also extended into the medial epicondyle, and one had additional comminution behind the capitellum.
Type II (Kocher-Lorenz). Only six patients had a sleeve type fracture of the capitellum, with equal gender distribution. All but one of these injuries extended beyond the capitel-Fig. 3
Lateral radiograph demonstrating the double arc sign.
Annual incidence of distal humeral articular fractures, by age and gender.
Male Female Occupation Number (%) Mean age (range) Number (%) Mean age (range) Professional 1 (7) 24 3 (7) 45 (28 to 64) Clerical 1 (7) 47 7 (15) 44 (36 to 54) Manual 3 (22) 27 (21 to 34) 1 (2) 56 Domestic 0 1 (2) 27 Student 8 (57) 17 (14 to 21) 5 (11) 17 (13 to 26) Retired 0 25 (56) 75 (62 to 91) Unemployed 1 (7) 34 3 (7) 26 (17 to 35) Total 14 (100) 23 (14 to 47) 45 (100) 57 (13 to 91)
lum. The trochlear ridge was involved in five cases, one of which also had involvement of the medial epicondyle (Ring type 2).
Type III (comminuted). This was the least common frac-ture, occurring in only three patients, aged 48, 81 and 91 years. They were more extensive than the radiographs sug-gested, all involved the trochlear ridge, and one extended to the medial epicondyle. None was associated with a radial head fracture.
Type IV (McKee). Of the 24 male patients, 13 (54%) had this coronal shear injury, which extended to the trochlea on pre-operative radiographs. It was also seen on pre-opera-tive radiographs in 36% (20) of the females. Six (18%) type IV fractures, all in women, had an associated elbow dislo-cation, compared with 1 (2%) in association with all other fracture types. The type IV injuries were more common with high-energy trauma (Table III). At operation most had more extensive articular damage according to Ring.9 Only Table II. Mechanism of injury, age (in yrs) and gender (unavailable in four
Mechanism of injury Number (%)
(range) Number (%)
Mean age (range)
Fall from standing 11 (50) 32 (14 to 62) 48 (90) 57 (13 to 91)
Fall from height 1 (5) 21 0
MVA* 6 (27) 28 (21 to 34) 4 (8) 44 (25 to 81)
Sport 3 (13) 22 (18 to 28) 1 (2) 15
Assault 1 (5) 17 0
Total 22 (100) 28 (14 to 62) 53 (100) 55 (13 to 91)
* MVA, motor vehicle accident
Table III. Descriptive radiological and Ring classification,9 age (in yrs) and gender
Classification system Number (%) Mean age (range) Gender ratio (M:F)
Descriptive radiological classification
I 37 (47) 51 (13 to 89) 0.2:1 II 6 (8) 37 (15 to 64) 1:1 III 3 (4) 73 (48 to 91) 0.5:1 IV 33 (41) 41 (13 to 82) 0.65:1 Ring classification 1 21 (26) 43 (13 to 91) 0.75:1 2 11 (14) 37 (14 to 65) 0.38:1 3 6 (8) 39 (17 to 63) 0.50:1 4 6 (8) 54 (14 to 78) 1:1 5 6 (8) 49 (17 to 81) 0.20:1 Capitellum only 29 (36) 53 (13 to 89) 0.26:1 Total 79 (100) 47 (13 to 91) 0.44:1
Table IV. Radiological and Ring classification,9 and mechanism of injury
Classification system Fall from standing Fall from height MVA* Sport Assault Descriptive radiological classification
I 31 0 5 1 0 II 3 0 2 1 0 III 2 0 1 0 0 IV 23 1 2 2 1 Ring classification 1 13 0 3 2 0 2 9 0 1 1 0 3 3 1 1 0 1 4 6 0 0 0 0 5 4 0 1 0 0 Capitellum only 24 0 4 1 0 Total 59 1 10 4 1
12 (36%) were the Ring9 equivalent (type I) of the McKee10
type IV injury, and the remainder were distributed across Ring groups 2 to 5 (Table V). Increasing age was associated with a greater degree of comminution, hence a higher Ring grade (Table VII).
Associated fractures of the radial head. A total of 19 patients (24.1%) had an associated fracture of the radial head (Fig. 5), with equal gender distribution (chi-squared test, p = 0.26). Five (26%) were Mason type I and 14 (74%) type II. There was no significant relationship between a fracture of the radial head and pattern of capitellar fracture (Table VI).
Other associated injuries. Two males had more complex soft-tissue injuries, one a Gustilo and Anderson16,17 type 3a
open Hahn-Steinthal fracture11,12 in a polytrauma case, and
the other an ulnar nerve lesion in a McKee type IV injury without dislocation. The neurological symptoms resolved spontaneously.
The accuracy of pre-operative radiological assessment of the injury. Of the 50 fractures that extended beyond the capi-tellum at operation, 33 were identified on pre-operative radiographs and 17 were considered as confined to the
capitellum (sensitivity 66%). None of the 29 fractures con-fined to the capitellum at operation was thought to be more extensive pre-operatively (specificity 100%). All 33 frac-tures thought pre-operatively to extend beyond the capitel-lum were confirmed to be complex articular surface fractures at surgical exploration (positive predictive value 100%), whereas, of the 46 thought to be isolated capitellar fractures from pre-operative radiographs, 17 were more complex at operation (negative predictive value 63%).
This study confirms that injuries to the capitellum are rare, with an annual incidence of 1.5 per 100 000 popul-ation over 13 years of age. The higher prevalence in the extent of the articular surface damage
Descriptive radiological classification (%) Type I Type II Type III Type IV
Ring classification 1 3 (8) 4 (67) 2 (67) 12 (37) 2 5 (14) 1 (17) 1 (33) 4 (12) 3 1 (3) 0 0 5 (15) 4 0 0 0 6 (18) 5 0 0 0 6 (18) Capitellum only 28 (75) 1 (17) 0 0 Total 37 (100) 6 (100) 3 (100) 33 (100) Fig. 5
Lateral radiograph of a shear fracture of the distal humeral articular sur-face with associated fracture of the radial head.
Table VI. Fracture of the radial head and Ring classification9
Ring classification (%) Mason classification of
radial head fractures
(%) Type 1 Type 2 Type 3 Type 4 Type 5
No fracture 23 (79) 13 (62) 8 (73) 6 (100) 5 (83) 5 (83)
Type 1 1 (4) 2 (10) 1 (9) 0 0 1 (17)
Type 2 5 (17) 6 (28) 2 (18) 0 1 (17) 0
Total 29 (100) 21 (100) 11 (100) 6 (100) 6 (100) 6 (100)
Table VII. Ring classification and age
Age in years Capitellum only (%) 1 2 3 4 5 Total
< 20 1 (3) 6 2 1 1 1 12
20 to 59 15 (52) 8 8 3 2 2 38
60+ 13 (45) 7 1 2 3 3 29
women over 60 years of age is probably related to the greater carrying angle of the female elbow and to post-menopausal osteoporosis.18
Distal humeral articular fractures are usually attributed to a fall on to an outstretched hand,19 and cadaver work20
has shown that capitellar fractures can be sustained by direct injury in flexion and indirectly through an extended radius, as in falling on an outstretched hand. Both mecha-nisms may produce a coronal shear fracture. Indirect trauma may be associated with a fracture of the radial head, as observed in one-quarter of our series. It is likely that these fractures represent the extreme of a spectrum of injury that includes minor scuffing of the capitellum, often observed with radial head fractures.21
Although the major types of isolated capitellar fracture were defined in the 19th century, the concept that they are often more extensive at surgery than initially appreciated from radiographs is relatively new. Accordingly, it is sug-gested that all fractures confined to the distal humeral artic-ular surface should be considered a subgroup of distal humeral fractures9,22 and based on specialist imaging or operation, and classified as such.9
Supported by low sensitivity and negative predictive val-ues from plain radiographs, we confirmed the observations of Ring et al,9 especially for type IV (McKee) fractures. Pre-operative computed tomography (CT) gives more informa-tion to assist in pre-operative planning, as many articular fractures cannot be approached by a standard lateral inci-sion. Ring et al’s9 study population were younger than ours, and that algorithm of internal fixation for all such injuries may not be appropriate in the elderly, who sustained the majority of these injuries in our series. Reconstruction may not be possible and hinged arthroplasty or non-operative treatment may be more appropriate in these patients.
This study was limited by its inevitably retrospective design owing to the low incidence of these fractures. Although the information from pre-operative radio-graphs reflects typical clinical practice, the data from the operation notes cannot be of the same quality as one would expect from a prospective study. We believe all patients, especially those over 60 years of age presenting with a fracture of the distal humeral articular surface, should have a CT imaging, but acknowledge that a
pros-pective study would be needed to determine whether this would alter their management.
No benefits in any form have been received or will be received from a commer-cial party related directly or indirectly to the subject of this article.
1. Cooper SAP. Fractures of the external condyle of the humerus. In: Lee, Cooper SAP, eds. A treatise on dislocations and fractures of the joints. London: Joseph Butler, one 1841:508.
2. Alvarez E, Patel MR, Nimberg G, Pearlman HS. Fracture of the capitulum humeri. J Bone Joint Surg [Am] 1975;57-A:1093-6.
3. Dushuttle RP, Coyle MP, Zawadsky JP, Bloom H. Fractures of the capitellum. J Trauma 1985;25:317-21.
4. Hirvensalo E, Bostman O, Partio E, Tormala P, Rokkanen P. Fracture of the humeral capitellum fixed with absorbable polyglycolide pins: 1-year follow-up of 8 adults. Acta Orthop Scand 1993;64:85-6.
5. Lansinger O, Mare K. Fracture of the capitulum humeri. Acta Orthop Scand 1981;52:39-44.
6. Ochner RS, Bloom H, Palumbo RC, Coyle MP. Closed reduction of coronal frac-tures of the capitellum. J Trauma 1996;40:199-203.
7. Scapinelli R. Treatment of fractures of the humeral capitulum using fibrin sealant. Arch Orthop Trauma Surg 1990;109:235-7.
8. Bryan RS, Morrey BF. Fractures of the distal humerus. In: Morrey BF, ed. The elbow and its disorders. First ed. Philadelphia: WB Saunders 1985:302-39.
9. Ring D, Jupiter JB, Gulotta L. Articular fractures of the distal part of the humerus. J Bone Joint Surg [Am] 2003;85-A:232-8.
10. McKee MD, Jupiter JB, Bamberger HB. Coronal shear fractures of the distal end of the humerus. J Bone Joint Surg [Am] 1996;78-A:49-54.
11. Hahn NF. Fall von einer besonderes varietat der frakturen des ellenbogens. Zeitsch.Wundartze Geburtshlefer 1853;6:185.
12. Steinthal D. Die isolirte fraktur der eminenthia capetala in ellenbogengelenk. Zen-tralb.Chir 1898;15:17.
13. Kocher T. Beitrage zur kenntniss einger praktisch wishctiger fraktur formen. Mitheil. a. Klin. u. Med. Inst. & Schweiz. Basal, reihe 1896:767.
14. Lorenz H. Zur kenntnis der fractural capitulum humeri (Eminentiae Capitatae). Dtsche Ztrschr. f. Chir 1905;78:531-45.
15. Mason ML. Some observations on fractures of the head of the radius with a review of one hundred cases. Br J Surg 1954;42:123-32.
16. Gustilo RB, Anderson JT. Prevention of infection in the treatment of one thousand and twenty-five open fractures of long bones: retrospective and prospective analyses.
J Bone Joint Surg [Am] 1976;58-A:453-8.
17. Gustilo RB, Mendoza RM, Williams DN. Problems in the management of type III (severe) open fractures: a new classification of type III open fractures. J Trauma 1984;24:742-6.
18. Grantham SA, Norris TR, Bush DC. Isolated fracture of the humeral capitellum. Clin Orthop 1981;161:262-9.
19. Keon-Cohen BT. Fractures at the elbow. J Bone Joint Surg [Am] 1966;48-A:1623-39. 20. Amis AA, Miller JH. The mechanisms of elbow fractures: an investigation using
impact tests in vitro. Injury 1995;26:163-8.
21. Newman JH. Radius fractures and damage to the capitulum humeri. Injury 1983;14:477.
22. Robinson CM. Fractures of the distal humerus. In: Bucholz RW, Heckman JD, Court-Brown CM eds. Rockwood and Green’s fractures in adults. Philadelphia: Lippincott Williams and Wilkins, 2006:1051-116.